facebookユーザー様からの情報です。


IOHK | Cardano whiteboard with Charles Hoskinson



英語ですが字幕は続きに収納しておきます。









hi I'm Charles house kansen Chu focued
of officer of input/output Hong Kong and
I'm here today to talk a little bit
about card ah no
so to understand card ah no first you
have to understand where Cardinal came
from so let's talk about the first
generation of cryptocurrencies so first
gen is Bitcoin and the problem that
Bitcoin was trying to solve was really
could we create a decentralized money
could we create some sort of token that
lives on some sort of decentralized
blockchain maintained by people all
around the world and that token would be
scarce and tradable so when Alice and
Bob want to send value to each other
there would be a mechanism for doing
that did that did not require a trusted
third party a trusted intermediary now
this was kind of a really cool and
interesting idea it had very old roots
starting from the 1980s and Beyond but
Bitcoin was really the first to bring
this all together and it was a
tremendously successful experiment after
just a few years Bitcoin not only
accrued thousands of users but also
started being worth real money tokens
went from less than a penny to actually
$1.00 to eventually $100 and right
around that time period we saw a huge
influx of people saying boy this is
really interesting however the issue is
that the transaction between Alice and
Bob has more than just the act of moving
money associated with it there's a story
behind that transaction there's terms
and conditions for example what if Alice
says the Bob I'll give you the money
if and only if you mow my lawn I'll give
you the money if and only if you repair
my roof this is a contract this is the
story so first generation of technology
wasn't really well suited for this every
single time someone wanted to make a
change to Bitcoin that have to build
another cryptocurrency or they'd have to
figure out how to install some sort of
cumbersome overlay protocol like master
coin or color coins so back in 2014
metallic pewter in myself and many
others came together and
launched the first second generation
blockchain and this is aetherium now
aetherium is kind of like when
javascript came to the web browser we
went went from these static simple pages
that were not terribly functional but at
least they did something to pages that
were fully programmable that this and
then enabled us to build the Facebook's
the Google's the Gmail's the experiences
we've come to know and love so atheria
brought a programming language to a
blockchain so this programming language
paradigm allowed smart contracts to be
written to have customizable
transactions so when Alice sends that
value to Bob all those terms and
conditions could then be embedded within
the transaction and it can be bespoke to
her particular needs
now this paradigm like Bitcoin also took
off and now aetherium is among the
largest of crypto currencies and has a
huge developer community unfortunately
just like Bitcoin we're starting to
enter into a new realm we're going into
the third generation where we've
realized that aetherium can't scale to
millions of users to billions of users
the theorem doesn't really have a good
developer experience at cerium and all
cryptocurrencies have really a bad
governance experience right now where
every single time there's a major
disagreement instead of finding a way to
resolve it we end up usually seeing
things like a theory emit a theorem
classic or things like Bitcoin and
Bitcoin cash furthermore there are big
sustainability problems in the space
namely after the ICO money is out for a
project that's funded this way or let's
say the venture capital runs out who
will fund and to grow the chain who will
fund to actually build out an ecosystem
these are big open questions so the
third generation is all about three
topics one is scalability two is
interoperability
and finally there's this notion of
sustainability
so the Cardno project is really our
philosophy our vision on how to solve
each of these categories in a way that
we feel it's going to inherit the best
features and lessons learn from
generation one in generation two but
also add a lot of new concepts and
technologies in space and this project
is built with some really good
principles namely - first off all the
science that guides the solutions to
these problems goes through some notion
of peer review so we go to conferences
we write proper scientific papers we
engage universities and all the
engineering we have the goal to
eventually implement as high assurance
code which means the same types of
techniques one would see with the
Shinkansen or a jet engine where when
the failure of the system results in
human death we can apply those
techniques to a to our protocols the
engineering and development so that we
have a much higher belief in the quality
of the code this avoids events like the
Dow the parity hack and other such
things as we've seen in the space so
let's go through each of these in more
detail starting with scalability so
scalability is kind of a loaded term and
it has a heck of a lot of meanings but
from a cryptocurrencies perspective you
can really think of scalability in three
different perspectives so one of
scalability in terms of transactions per
second so you'll often hear people say
well bitcoin has seven transactions per
second or a theorem has 10 or 20
transactions per second and this new
protocol has 200 transactions or 300
transactions per second this notion of
how many transactions are able to get
into a block within some finite period
of time but it's not the only thing that
you have to concern yourself with
transactions carry data and
as you get more transactions you require
more network resources so there's also
this notion of bandwidth or network
where for a system to scale if it's
going to go to millions and billions of
users that system could require hundreds
of megabytes to gigabytes per second of
bandwidth to be able to support all the
data flowing through it this is their
common in the enterprise world but not
quite where we need to be in the
peer-to-peer world and then finally
there's this notion of data scale so
block chains store things hopefully
forever and so every time you put a
transaction in regardless if it's
relevant or not it ends up in the log
and as you have more and more
transactions per second you need more
and more data and as a consequence block
chains will grow from megabytes to
gigabytes two terabytes to petabytes and
potentially even exabytes again this is
okay in the network world but when we
talk about a replicated system whose
security model relies upon each node
having a copy of the blockchain this is
not tractable for consumer hardware
devices so Cardno
what we're trying to do is figure out
ways to solve these problems in a very
elegant way namely that as we add people
to the network we naturally get more
transactions per second we get more
network resources and eventually we'll
get more available overall data storage
without compromising our security model
so let's talk a little bit about some of
the innovations we've already brought
namely with the throughput we have
developed a peer reviewed white paper
for our provably secure proof of stake
protocol
called or Boris and Ora Boris is among
one of the most efficient consensus
protocols in the cryptocurrency space
and it's the first to actually be proven
secure in a very rigorous cryptographic
way the magic of hora Boris is that it's
been designed in a modular way and it's
been designed with future proofing in
its DNA so namely how hora Boris works
his aura Boris breaks the world in two
epochs it takes a look at the
distribution of tokens and grab it from
a source of random numbers that's able
to hold an election and create slot
leaders now these law leaders
functionally do the exact same thing
that a miner would do in Bitcoin so this
is basically the same as a person who
discovers a block wins a block and
Bitcoin but the difference is that it
doesn't require the extensive
computational resources that Bitcoin
requires to construct a block and as a
consequence this system is considerably
cheaper to run even though we still have
similar security guarantees that Bitcoin
currently enjoys so it's a major
advancement but the other really
interesting thing is that these slot
leaders don't have to just maintain a
single block and a single chain they can
actually maintain other blocks and other
chains because the cost of constructing
a block is so low it's actually now
tractable to talk about consensus over a
range of block chains instead of a
single chain furthermore ethics perhaps
could even be run in parallel so instead
of having one epic run and then another
epic run one going to develop a system
using or Boris we're epics run in
parallel and transactions are
partitioned accordingly what this
effectively means is as you gain more
users and your users gain more
capabilities these slot leaders will be
able to maintain more types of block
chains and also run
transaction processing for blockchains
in parallel this is a major advancement
the other cool thing is that were Voris
has very rigorous security standards in
terms of its theoretical foundations as
well as its implementation so as a
consequence as we develop new
capabilities for the protocol these
protocol ''tis are somewhat composable
meaning that these capabilities will
also be secure whereas other systems one
has to prove these things on a case in
case basis in some cases make major
modifications or changes to their system
in addition to this we intend on Ora
Waris actually being quantum resistance
sometime in 2018 hopefully where when
the slot leader signs these blocks
they'll be using a quantum resistant
signature scheme so we can get even more
future proofing into the system so
that's kind of the first view is how do
we construct a way of maintaining the
network that doesn't cost three hundred
thousand dollars an hour which is what
Bitcoin currently costs at the current
market rate of five thousand per coin
and how do we build a system that allows
us to go parallel and also allows us to
potentially maintain multiple chains
concurrently this is the heart of war
Boris now as we mentioned before one of
the most important things when one
develops new cryptography is to make
sure that this cryptography is developed
in a very rigorous peer-reviewed way so
Ora Boris actually was accepted at
crypto 17 where our team went there and
to present it and future versions of the
protocol continue to go through more
rigorous peer-review giving us a high
assurance that the conceptual design of
the system is correct
the other side of it is we're actually
modeling a formal specification of hora
Boris using side calculus which is a
wonderful formal modeling language
that's machine understandable that
eventually we're going to be able to
connect to the Haskell code in our
github repo and actually show that we've
correctly implemented the protocol this
is a standard that actually does not
exist in our space and we're very
excited to be the first to bring it to
the space but that's not the only thing
required for scalability you also need
the ability to move large amounts of
data at
same time as our network grows from a
few hundred transactions per second to
thousands to tens of thousands to
hundreds of thousands of transactions
per second you cannot maintain a
homogeneous network topology in other
words you cannot have a situation where
every node has to relate every message
because there are will be nodes that
don't have those capabilities especially
as they grow so we're actually looking
at a new type of technology called Rena
and this stands for recursive internet
work architecture and so Rina is
basically a new type of structuring
networks using policies and clever
engineering principles mostly conceived
by John Day who's based out of Boston
University and the goal here is to build
a heterogeneous network that gives you
similar types of privacy guarantees
transparency guarantees scalability
guarantees that you would expect to get
from network protocols like TCP IP but
do so in a way where you can actually
reason about how the network is going to
compose in a formal capacity so in other
words Reena is a major step forward that
will give us a way to very naturally
tune and configure Cardno as it grows
from hundreds to thousands to tens of
thousands of transactions per second and
something that will very seamlessly
connect and interoperate with tcp/ip
we're very excited to bring this in part
to the Cardona ecosystem in 2018 and in
completely around 2019 and we think it
will be a big solution to a lot of the
network overhead issues that we have
finally we have this notion of date-ish
scaling so this is among the hardest of
the problems and it's one that we're
still examining and looking at but the
reality is that not everybody requires
all the data the transactions that Alice
sends the Bob are not necessarily
relevant to even Jayne and Bill they're
only relevant from the
context that these people are able to
infer from them the tonal ownership the
money and know that the tokens they
receive are legitimate and correct so
you have techniques like pruning you
have techniques like subscriptions and
some form of compression which one
applied in a setting in a very
intelligent way can actually reduce
substantially the amount of overall data
one user has to have there's also the
idea of partitioning partitioning is
where a user doesn't actually have a
full copy of the blockchain rather they
have some chunk of the chain as opposed
to many other chunks and there's some
notion that you can put all of these
things together so what are the goals of
the Cardinal project is to study this in
a very rigorous way and to come up with
new blockchain architectures as they
become necessary that will allow people
to have much smaller amounts of data but
overall still get the same level of
assurance and security that the
transactions they receive the
computations that are performed are
correct so part of this research also
involves some of our ideas about side
chains and we'll get to that actually
when we get to the interoperability
section but as a brief mention all a
side chain really is is two components
first it's the notion of creating a
compressed representation of a
blockchain and it's a notion of creating
interoperability between chains ie a way
of translating transactions between
chains so some of the proofs that we've
come up with with some of our recent
sidechains research give us a lot of
hope that we can create compress
representations of a blockchain as it
grows from gigabytes to terabytes
to petabytes and that these proofs can
give us high level of certainty that the
transactions we're seeing the history
we're seeing is correct even though
these proofs are quite small megabytes
two kilobytes so this is kind of our
solution to data is to approach it and
in both a pragmatic way where we start
restricting in some
cases what people see in a very
intelligent case-by-case basis we
partition things where and when we can
and we find out really intelligent
cryptographic ways to compress history
which give us the same level of
confidence even though we don't have all
of it but do all of this in a way where
we don't remove the original replicated
security guarantee that Bitcoin has has
endowed the space with the other
fortunate thing is well TPS tends to
grow quite a bit and network resources
also tend to grow quite a bit storage is
still relatively cheap and available so
we believe that the data scaling side of
Cardno will be something that we don't
have an urgency to resolve until late
2018 to mid 2019 the research has just
started at a University of Edinburgh and
we intend to continue it and we believe
we will have our first approaches
probably by the middle part of 2018 and
eventually have a total solution to this
problem by the end of 2019 for the
system so this is kind of the first
component let's build a protocol that
allows us to scale up the TC
transactions per second so as we gain
more people into the ecosystem we get
more throughput let's create a network
stack developed by one of the putter
early internet pioneers that has learned
from some of the issues that tcp/ip a
has and other systems and is a bit less
fragile and a bit easier to configure in
tune so that it can handle a large
heterogeneous network based upon
real-life scenarios and let's come up
with some clever ways of handling the
use of data in our system without
compromising our security guarantees and
also when a partitioning is adopted so
much like BitTorrent you actually gain a
lot more resources for overall storage
so if we don't have to replicate the
database rather we can distribute it
there is some hope that we can build a
large distributed file system for Cardno
and thus actually have total available
petabytes of data even though any
particular user only has to have a small
amount so that's the first pillar this
notion of scalability but the third
generation demands to more and therefore
for the Cardinal project we will do
these things first
there's this idea of interoperability so
interoperability is this idea that there
will not be one token to rule them all
so sorry Bitcoin you're going to have
many networks like aetherium and Bitcoin
and ripple and legacy systems like the
traditional bank networks running on
older protocols like Swift and older
settlement networks like ACH and so
forth and these systems all speak their
own languages and these systems all have
their own business logic and rules so
currently as it stands it's very
difficult for aetherium and Bitcoin to
understand each other substantially more
so for the old banking networks which
have the added requirement of metadata
and attribution to the transactions so
the problem is that if you don't have a
single standard and if you don't have a
canonical way of communicating with
these systems then you run into a
situation where value gets very
fragmented so regardless of how
decentralize any one of these particular
ecosystems happen to be the kingmaker
will be the small on and off boarding
hubs that control the movement of value
between these systems currently we are
seeing those as exchanges but there are
others who could come and these
exchanges are very fragile they're
subject to being hacked they're subject
to draconian regulation occasionally
they get shut down because of unwise
regulatory policy or in some cases wise
regulatory policy and this is really not
a good situation to be in for a
supposedly decentralized permissionless
ecosystem to have a small group of
actors control whether one can convert
their value from one system to another
system furthermore when people do
business in this world if these
businesses are regulated or even semi
regulated they usually do have to
interface and interact with the
traditional financial world for example
let's say that you're a cryptocurrency
company and you issue an e rc 20 token
okay so that ERC 20 token let's see you
have a crowd sale and you raise millions
of dollars worth of ether and as a
actual company with a bank account in a
legal jurisdiction you begin selling
that ether and deposit millions of
dollars into your bank account
well then Bank is a regulated entity
there and the first question they're
going to ask is where did you get these
millions of dollars from and the bank
will then ask you to to explain to
provide some data and details you'll say
oh well I I had a crowd sale and I sold
an e rc 20 token and I got a bunch of
ether they say okay well who did you get
it from who are your customers and you
respond people over the Internet and
unfortunately that's really not a good
answer and this entity has a regulated
business entity that has to file
suspicious activity reports that has to
deal with people on their side like the
Treasury Department or maybe European
Union and so forth these entities feel
that this is a very risky proposition
and this is the unfortunate reality that
we tend to live in in our world we have
fragile inter links throughout crypto
currencies as well as the legacy
financial system and there's really no
way to escalate transactions in a very
natural way so that when one wants to do
business with the legacy world that
metadata that attribution the compliance
information that one would want to have
is not present and as a consequence
anybody doing business here
automatically becomes a high-risk
business this is an unfortunate
situation so the idea of a
third-generation cryptocurrency with
respect to interoperability is a
cryptocurrency that has the capacity of
being able to understand and watch other
cryptocurrencies a cryptocurrency that
when it sees aetherium an event
occurring here can actually verify if
that event is true or false
for example if Alice says she has ether
and sends ether to Bob this
cryptocurrency ought to be able to
actually know that that's a legitimate
transaction if it's something in the
concern of the third-generation crypto
currency so cross chain transfers are
reliable and they should be able to do
this without needing a trusted third
party that's the single most important
thing because we want to create an
Internet of blockchains an internet of
value that flows around just as easily
as Bitcoin flows around or ether flows
around we want to move cross chain so
the first component of that is to have
some notion of side chains though this
is not a new idea
atomic cross chain swaps or side chains
these things have been around for a long
time they've been proposed since as
early as 2012 perhaps even sooner but
the basic concept is that there is some
way of structuring information from one
chain to another change such that when a
transaction is sent that compressed
structuring of information gives you the
ability to know if that transactions
legitimate or not in other words the
person sending it to you actually has
that value and also this value is not a
double spent it's a very important
concept the other side of it is that you
have to have the ability to do this in a
very compressed way there are over a
thousand cryptocurrencies in use and
cryptocurrencies are becoming larger and
larger so you can't say well the only
way to understand the other system is to
have a copy of the entire blockchain of
the other system this is not a scalable
solution you have to be able to look at
these systems in a very compressed way
so Cardno has started its sidechains
effort we recently published a paper
with Andrew Miller Danica's Andros and a
glassy osseous and this paper contains a
very well thought out approach for how
to generate proofs in the proof-of-work
world called non interactive proofs of
proof of work and we're very hopeful
that this approach can also be adapted
in the proof of stakes setting and then
these two things together combined with
some clever engineering should allow us
to have a pretty deep and detailed
understanding of what's going on amongst
other crypto currencies which are here
to stay however this is only part of the
story even if we can create a utopia
where all block chains can talk to each
other and be seen and heard and they go
from blind deaf and dumb
two very intelligent and braavos the
issue is that this world is still
incompatible with this world principally
because of three factors one is the
notion of metadata two is the notion of
attribution and three is the notion of
compliance so metadata is the story
behind a transaction it's not that
you've spent fifty dollars that matters
it's where did you spend it what did you
spend it on to whom did you give it to
these types of things
that's metadata it's really not well
provisioned in the cryptocurrency space
however it's the bread and butter of the
legacy financial world there's an
enormous amount of value in the metadata
of transactions and in some cases the
metadata of transactions allows
transactions to be put into a hierarchy
of risk certain types of transactions
are tremendously risky for example if
there's a wire transfer between two
American banks based in New York amongst
large companies this is very common and
happens all the time if there's a wire
transfer from a small American company
to a small Russian institution which in
turn then goes to Iran which in turn
then goes to South Africa this is a
totally different scenario even if the
amounts are the same the amount of hops
the people who have touched it the
nature of the organizations how long
they've been doing business this is all
metadata the problem is that metadata is
incredibly personal it's incredibly
private and it has a big difficulty in
the cryptocurrency space because all the
transactions here are permanent they
live on a blockchain they're transparent
and the minute that we would attach
metadata to a transaction here we'd run
into a situation where we could be
potentially exposing very sensitive
information to the general public so
what are the goals of the card ah no
project is to figure out where when and
how we can put metadata on a blockchain
and benefit from some of the factors
that we really care about such as audit
ability as well as immutability and time
stamping because tempering with metadata
is a very very risky thing it's very
important in cases to have this but at
the same time posting metadata to a
blockchain in a responsible way
perhaps encrypted or perhaps with a
special scheme that only allows certain
people to see it whereas others cannot
attribution is about identity it's about
the naming of actors involved in
transaction it's a subset of metadata
but it's so important it deserves its
own consideration it's not just good
enough to know the story behind the
transaction you have to know where the
money came from and where the money's
going that's Alice to Bob JPMorgan Chase
the Wells Fargo and so forth this is so
incredibly important that we've decided
to try to construct a way where if one
desires to add attribution to a
transaction it can be done in a pretty
streamlined and easy way but the
difficulty is establishing a web of
trust or some sort of identity hierarchy
one of the reasons why we live in an
internet with passwords and usernames is
because we actually don't have a good
way of identifying the people on the
Internet it would be great if everybody
had a public key and there was an easy
way to distribute these things and an
easy way to verify that Bob's public he
was actually Bob's public key and this
was one of the goals of the PGP a
project it was never realized as a
consequence the second consolation prize
that we've had to deal with on the
Internet has been this terrible dystopia
where one has a username and a password
usually easy to guess usually easy to
hack and usually reused amongst many
different websites and causes certainly
a lot of problems so the same problem
there is incumbent here the difference
is that cryptocurrencies are factories
for cryptographic credentials they give
you a place to store public keys they
give you a place to store a web of trust
and a whole bunch of cryptographic
mechanisms that go above and beyond what
one would see in a normal web site
crypto currencies are basically
factories for cryptographic credentials
unlike normal websites are unlike the
current internet crypto currencies give
you the ability to organize manage store
the keys and develop all kinds of webs
of trust as well as an even more
advanced layer cryptography which has
been invented over the last 20 years so
part of the goal of the Cardinal project
is to start exploring how we can use
these things that we're currently using
for storing and saving our money to
actually also identify ourselves when
how we want to do and the hope is that
this can then be used when people are
required to give attribution of
transactions for example perhaps when
they send value to an exchange or back
they can do so in a very graceful and
easy way now the final component is this
compliance component and compliance is a
construction of things like kyc which
stands for Anno your customer things
like AML it sends for anti money
laundering and things like anti
terrorist financing ATF and basically
they're all the same notion of there's a
transaction that's occurred what do we
know about this transaction such that we
can say it's a legitimate transaction
this is something that is not really
considered in the crypto world but it's
the bread and butter of every single
financial institution whether they're in
exchange they are a bank or anybody
who's in a position to be a money
service business where they're handling
money on behalf of somebody else there's
very harsh and strict global regulation
in standards for how anti money
laundering and know your customer and
compliance needs to work in general so
with respect to this our hope is that we
can find a healthy balance where once
we've distributed these cryptographic
credentials and once we have provisions
for metadata that these two factors can
be put together in a very creative way
on a case-by-case basis and a voluntary
basis so that when somebody in the
crypto world wants to do business in the
legacy world they have an ability to
escalate the transaction from a standard
crypto currency transaction to one that
a bank would actually recognize and feel
very comfortable with for example the
scenario that we gave in the beginning
of this presentation about this ERC 20
token crowd sale one could entirely
imagine the idea of saying that the only
way one can send ether to this
transaction would be if it stamped with
some metadata accessible to the person
being sent to and it has some identity
information that's transitive such that
when they cash out and go to the bank
they can disclose that to the bank but
the key here is doing this in a way that
protects privacy and doing this in a way
that doesn't necessarily make people
custodians of the data as we've seen
with Equifax hack and other such things
custodianship of personally identifiable
information is quite problematic so a
big goal of the Cardinal project is to
explore this side of the space using new
cryptography using optional metadata and
also using things like trusted Hardware
for example which give us all kinds of
capabilities from very secure ways to
store credentials to the ability to
provide guarantees that data has been
destroyed after a period of time to
things like geo tagging for example so
this is the interoperability side and if
we're successful with this we can really
think of Cardno is that glue that can
facilitate the Internet of blockchains
where Bitcoin can stay Bitcoin aetherium
can stay as aetherium ripple can stay as
repo and the banks don't have to change
much but Cardano provides that necessary
bridge that isn't centralized and isn't
fragile rather it's a large
decentralized network and it really
ushers in this great new era of
interoperability in the last two
sections we talked about scalability and
interoperability respectively but the
third pillar of a third generation
blockchain and what a Carano aspires to
be and probably one of the most
important of the three pillars is this
idea of sustainability and you could
break sustainability into two components
first there is the idea of how do we pay
for things
so crypto currencies are not companies
even if one is to say a crypto currency
or a token as a security it's still
somewhat decentralized it's still
infrastructure it's much more like roads
it's much more like tcp/ip their
protocols and so when you talk about a
protocol that's open-source
the idea is to minimize the operational
cost of that protocol so things like
putting tolls on the protocol or
intellectual property on the protocol
even if it's well intended for example
to fund a foundation are likely to be
less competitive than completely open
completely free protocols so as a
consequence it's very difficult to
figure out exactly how one ought to
maintain these systems pay for these
systems in the long term so certain
proposals such as a patronage type of
model where a corporation decides to
volunteer its developers may work in the
short term but the challenge is that
those developers end up having a huge
amount of influence over the evolution
and growth of the system so one can
argue that a patronage model will lead
to centralization of power into the
hands of a few companies that want to
modify the protocol in a certain
direction for example right now the w3c
is having a large debate about bringing
DRM into the browser and we've already
seen the e FF resigned from the w3c
because they feel that larger companies
in that effort are influencing this
process towards the benefit of large
content distributors as opposed to a
free and open web so patronage may not
be the best way I SEOs are also an
interesting way and they're like a quick
jolt of energy they provide a lot of
capital and if there's good governance
behind that capital and good people
behind the capital certainly can result
in the creation of a great product a
great protocol but the problem is that
an IC o---- is an indefinite event or a
continuous funding event rather it's a
large flood of money in the very
beginning to do something but no matter
how large that pile of money is it's
finite and
eventually run out so the first thing is
can we construct a system that has a
Treasury a Treasury is where a
blockchain is able to print money and
put some of that money into a
decentralized bank account
and that decentralized bank account is
funded through inflation so the first to
pioneer this model was - if you look at
Bitcoin every time a block is produced
it started with 50 coins per every block
and then it decreased by factor of two
every four years and now we're down to
12 and a half all of that goes to the
miner with the - model instead of having
all of it go to the miner some of it
actually goes into this decentralized
bank account then there is a democratic
method to vote on funding proposals so
what Bob can do is Bob can submit up a
ballot to the Treasury and then the
token holders can vote in if there's
high enough threshold if there's a high
enough amount of voting
Bob's ballot will be approved and the
Treasury will open up in paid off so
this abstract model is actually
incredibly robust because it has a way
of being refilled on a continuous basis
it's directly proportional to the
overall influence of size of the
currency so as the currency grows it has
more and more resources available which
in turn can be spent to grow the
currency so there's a positive feedback
loop and also it has a democratic
process a Democratic participation
connected to the system that allows the
stake holders in the system to start
having discussions about priorities
namely what ballots ought to be funded
for example Bob can say I'm a developer
and I want to sustain the system and
submit a ballot and then Alice can say
I'm a marketer and I want to market the
system and create content videos and now
the token holders get to choose what do
they think is a higher priority further
development or marketing or perhaps a
scientist wants to write a new protocol
to improve the existing protocol maybe
it's a new privacy primitive for example
or maybe it's a bit more security with
the network stack or a bit more
anonymity with the network stack so
instead of going to the National Science
Foundation
the European Union and trying to get a
traditional research grant scientists of
the future may even be able to use the
Treasury model to propose funding for
research or perhaps you're an
application developer adapt developer
maybe you want to bring state channels
to aetherium under the current model one
has to do an ICO or something along
those lines issue an unnecessary token
that really doesn't provide any utility
above and beyond what a ether provides
but do so to raise capital in order to
be able to bring this innovation into
the system instead of having that and
segregating value away from the
principal token the token holders now
have a mechanism where they actually can
provide funding for this dap and that's
a really exciting proposition the
challenge however is that the
construction of such a system is a
tremendous endeavor and requires quite a
bit of deep thought first off there has
to be a proper and fair voting system
second there has to be incentives to
vote incentives to participate above and
beyond some notion that this system is
going to to be a common good there's a
kind of a tragedy of Commons notion
there third there has to be an easy way
to submit ballots but do so in a way
where ballots actually the reasonable
ballots have higher precedence and
priorities over absurd ballots and do
this all in a way that's completely
decentralized and do this all in a way
that is somewhat distributed and doesn't
require centralized governance so a
Treasury is quite hard - had a first
example of it they've had tremendous
success as a project as a consequence
so at IOH cave we are quite interested
in this model we think this is one of
the single most important things about
sustainability because it's leaving
something behind embedded within the
system itself that allows the system to
pay for its own bills yet it is as I
mentioned quite a bit of research so
we've already started the process we
first started examining the voting
system and we're looking into using a
modification of liquid democracy
and then combining that with an
incentivized Treasury model that we have
developed with a joint partnership with
researchers at Lancaster University as
well as some of our personnel that come
from Ukraine and our hope is develop a
reference Treasury model that's
abstracted from any cryptocurrency it's
just a module that can be plugged into a
cryptocurrency such as the theorem
classic or Cardinal the hope is that the
system will have a fairly sophisticated
way of balancing the needs and
incentives of token holders as well as
the needs of those seeking funding in a
reasonable way now this is an inexact
science it's one that requires quite a
bit of iteration and this is why we're
trying to construct this system in a
modular capacity so that the treasury
system can be upgraded independent of
the protocol itself if necessary so this
is something we're very excited about we
should be publishing a paper fairly soon
prying quarter four of 2017 and our hope
is to roll out the first generation of
the reference treasurer model into the
card on a protocol by about the middle
part to the end of 2018 depending upon
how long the research takes but there's
another side to sustainability it's not
just about how do we pay for things it's
also about where should we go and this
is an even bigger meta question
so cryptocurrencies are living creations
they're not static in that once you've
written the code it's finished it's done
there's no notion of 2.0 you have to be
able to change them as technology
changes use cases change or new
innovations come out that you can
benefit from tremendously and generally
you do this either with soft forks or
hard Forks but these are Forks
nonetheless meaning that you have to
change something the problem is that in
current cryptocurrencies generation 1 &
2 there is no canonical way of deciding
which fork is proper amongst many that
have been proposed and as a consequence
this results in a situation where
eventually irreconcilable differences
develop and we see the chain break apart
we seen this with Bitcoin and Bitcoin
cash we've seen this with etherium and
etherium classic and it will only
continue as these systems grow on scale
and value so what we have to do is argue
by analogy are there any social systems
that humanity has constructed which
while they may be controversial have
still maintained a degree of stability
and update ability constitutions are the
closest notion that we have that's the
highest law of a country a constitution
doesn't change too much especially
constitutions like the American
Constitution and they're somewhat
difficult to change but when they do
change there's universal consensus post
change that they're going to follow this
we don't see a fracturing of the country
into two Americas or two UK's or
something like that we kind of get
consensus around it why because it's a
slow deliberate process one that people
agreed to follow at least in principle
so if we were to treat protocols as
constitutions one ought to be able to
amend the Constitution one ought to have
a process to do so because we're
engineers and we like reusability
our hope is to use the very same types
of mechanics
that are used in the Treasury system for
approving ballots to actually being able
to consider card ah no improvement
proposals so Alice or Bob the developer
can propose a sip and then there is
going to be a process which Alice and
Bob can follow that allows the network
to vote on whether this should be
ratified or not this should be a slow
systematic deliberate process that takes
time and effort with increasingly higher
thresholds prior to eventual adoption
now the first generation of this
technology is more about process and
mechanism it's a meta consideration that
lives outside of the network but relies
on tools within the network for example
the ability to do voting by stake weight
later versions we're going to explore
the idea of converting a card ah no
improvement proposal into something
that's machine understandable similar to
a formal specification of a protocol
there is this notion that maybe we can
actually specify a cryptocurrency in a
way that the cryptocurrency actually
understands its own design if this is
possible then it may be possible to
actually verify if a client is following
a specification in other words when
Alice and Bob create their wallets to
connect to card on oh there's this
machine understandable canonical notion
of Cardinal ought to be configured and
their clients will actually be able to
self accredit and verify if they're
actually following the protocol or not
so this is kind of generation two and
three this idea of taking a social
process and eventually mechanizing that
social process in ways that give us a
protocol that lives above the particular
software and we're very hopeful that we
can use emerging techniques from
programming language theory and formal
verification to pursue these ends so the
first micro exam
of this are going to start materializing
with the formal verification of smart
contracts which is a very important
topic for reducing bugs and ensuring
that smart contracts are stable and
reliable and this is an area of research
that we're pursuing very vigorously at
i/o HK and we intend on having several
publications come out throughout 2018
our hope is that those techniques can
then be raised a level and we can begin
having protocol level descriptions of
how Cardno ought to work as a full
cryptocurrency stack and then eventually
we can pursue formal verification that
clients are pursuit are connected to a
specification in the short term however
the card ah no improvement proposal
process will be completely written up by
quarter 1 of 2018 and IOH gate will
begin following that process as it
upgrades and iterates the Cardinal
protocol and once our treasury voting
system is in place we will make special
accommodations so that every card on
owen provement proposal can eventually
go through a process of voting with
higher rigor on hard Forks and less
rigor on soft Forks but nonetheless
there's still some notion of this we
already have an upgrade system built-in
if one goes to Cardinal Docs there's a
specification on Cardinal Docs calm for
how that system works and later on that
system will be superseded by our new
voting system that we intend on rolling
out with the Treasury so this is
sustainability in a nutshell it is our
view on how we should responsibly pay
for things and our view of where should
we go as protocol start there's usually
a lot of founder vision and a lot of
initial philosophy but as more people
come into the ecosystem as more hopes
and dreams get invested into the
ecosystem eventually these ecosystems
grow beyond their founders and so it's
incredibly important to have a process
for changing the protocol in a very slow
deliberate methodical way that's
inclusive as possible and as resistant
as possible to any particular power
brokers ability or machinations
so that's Cardinal Cardinal is a
third-generation protocol it's built
with peer review it's built with high
assurance software standards it's built
in Haskell my fav
programming language it's built by a
large international team and are very
well funded in Iowa case committed to it
until 2020 and it's built to be
sustainable interoperable and scalable
our view of how do we get
cryptocurrencies from the first million
to the first billion thank you so much
for watching and I hope to hear from you
guys soon
 
タグ ：
#動画